| 引用本文: | 程超,张汉敬,景志敏,等.基于离群点算法和用电信息采集系统的反窃电研究[J].电力系统保护与控制,2015,43(17):69-74.[点击复制] |
| CHENG Chao,ZHANG Hanjing,JING Zhimin,et al.Study on the anti-electricity stealing based on outlier algorithm andthe electricity information acquisition system[J].Power System Protection and Control,2015,43(17):69-74[点击复制] |
|
| 摘要: |
| 为了解决以往依靠用户每月电量变化、台区线损异常确定窃电嫌疑户等传统方法时效性差、准确率低，现场排查发现窃电行为又效率不高的问题，利用当前远程集抄海量数据为传统反窃电理论研究和高效反窃电的实现尝试探索一种新方法。全面分析窃电手段及原理，绘制实施窃电手段的鱼骨图，由实际案例统计分析确定研究重点为欠压、欠流窃电法。根据高供高计和高供低计用户计量装置接线特点、电压电流值的规律以及离群点理论确定窃电判定算法，并做出基于距离的离群点检测法判定窃电的流程图。 通过实例验证提出的算法和窃电户筛选流程能够完全甄别出窃电用户，为监控人员利用用电信息采集系统的海量数据精确及时地进行有效反窃电分析提供了新思路。 |
| 关键词: 窃电 鱼骨图 离群点 用电信息采集系统 算法 |
| DOI：10.7667/j.issn.1674-3415.2015.17.011 |
| 投稿时间：2014-10-27修订日期：2015-01-13 |
| 基金项目: |
|
| Study on the anti-electricity stealing based on outlier algorithm andthe electricity information acquisition system |
| CHENG Chao,ZHANG Hanjing,JING Zhimin,CHEN Ming,JIAO Lei,YANG Lixin |
| (State Grid Qingdao Power Supply Company, Qingdao 266000, China) |
| Abstract: |
| In order to solve the problem that traditional methods such as screening the electricity stealing suspect users rely on monthly power consumption change, abnormal line loss is of low efficiency and low accuracy rate; and the field investigation finds that act of electricity stealing is inefficient. This paper explores a new method in traditional anti-electricity stealing theory study and efficient anti-electricity stealing implementation using the massive data of current remote centralized meter reading, analyzes electricity stealing means and principles, draws a fishbone diagram of electricity stealing means, determines this paper’s research emphasis are under-voltage electricity stealing method and under-current electricity stealing method by statistical analysis of actual case. Electricity stealing decision algorithm is determined based on metering device wiring characteristics and voltage/current value law of high voltage power supply high voltage measurement users and high voltage power supply low voltage measurement users, outlier theory, an electricity stealing determining flowchart is drawn based on distance-based outlier detection method. Examples validate that the proposed algorithm and electricity stealing user selection process can fully identify electricity stealing users, providing a new idea to monitor staff for the accurate and timely anti-electricity stealing analysis with massive data of electricity information acquisition system. |
| Key words: electricity stealing fishbone diagram outlier electricity information acquisition system algorithm |
